Many toxins kill cells by the metabolic generation of chemically reactive intermediates that in turn produce alterations in cellular molecules, compromising cell structure and function, leading to loss of viability. Many chemical and biochemical changes are observed in such studies, not all of which are involved critically in the initiation of damage. Nevertheless, three major classes of tissue alteration are recognized: alkylation, oxidation, and peroxidation. A difficulty in the study of oxidant stress-induced hepatic necrosis has been the lack of an animal model. Although the mechanistic involvement of reactive oxygen species has been proposed for tissue damage initiated by a number of drugs and by reflow following ischemia, much of what is known about the mechanisms through which reactive oxygen species kill cells has been developed in studies conducted in vitro, particularly in isolated or cultured rat hepatocytes. We have found that diquat produces centribolular necrosis in male Fischer-344 rats and several lines of evidence indicate that this necrosis is mediated by the generation of reactive oxygen species. This model has been studied further in an effort to understand the manner in which reactive oxygen species and oxidant stress kills cells in vivo and to develop quantitative criteria for distinguishing those examples of cell death in vivo that may be mediated by oxidant mechanisms from those in which the lethal effects ar expressed through other mechanisms. Investigations of the diquat model of acute hepatic necrosis have demonstrated some important differences from the hypotheses generated from studies in vitro, particularly the role of shifts in cellular thiol/disulfide ratios and the depletion of protein thiols. These studies of the diquat model have revealed an apparently critical role of lipid peroxidation in the expression of reactive oxygen injury in vivo, and that the availability of chemically reactive chelates of iron appears to be an essential determinant of peroxidation damage and cell death. The studies described in this proposal are designed to apply chemically specific methods of analysis to the investigation of the extent to which lipid peroxidation is responsible for diquat-induced hepatic necrosis in vivo and the role played by chemically reactive iron species in these processes. These methods and approaches and the quantitative criteria developed in these studies will be applied to a critical examination of the hypothesis that oxidant stress mechanisms contribute to hepatic necrosis initiated by acetaminophen an by carbon tetrachloride. Because the latter two agents are such important model hepatotoxins, the proposed studies are directed at issues that are fundamental to our understanding of the mechanisms of cell death in vivo.